I am a Research Engineer at the French Alternative Energies and Atomic Energy Commission (CEA), affiliated with the List Institute
My research focuses on information extraction using few-shot learning for event extraction and domain adaptation. This involves deep learning techniques in NLP, including sequence modelling, structured prediction, graph neural networks, and recent meta-learning methods such as prototypical networks and memory-augmented networks.
My professional expertise and interests also encompass deep learning, natural language processing, computer vision, and generative AI. I am currectlly working on multimodal large models for information extraction and understanding.
Navigating the Challenges of Vision-Language Models Introduction Vision-Language Models (VLMs) represent a significant advancement in multimodal AI, enabling systems to process and understand bot...
Few-Shot Event Argument Extraction: A Meta-Learning Approach Introduction In natural language processing, event argument extraction is a crucial task that involves identifying and classifying the...
Information Extraction in Practice: A Comprehensive Guide Introduction Information Extraction (IE) is a crucial task in Natural Language Processing that involves automatically extracting structur...
Understanding Attention Mechanisms in Deep Learning Introduction Attention mechanisms have become a cornerstone of modern deep learning architectures, particularly in Natural Language Processing ...